Automatic Preprocessing Technique for Detection of Corrupted Speech Signal Fragments for the Purpose of Speaker Recognition.
Konstantin SimonchikSergei AleinikDmitry IvankoGalina LavrentyevaPublished in: SPECOM (2015)
Keyphrases
- speaker recognition
- speech signal
- speaker identification
- gaussian mixture model
- vector quantization
- speaker verification
- speech recognition
- noisy environments
- automatic speech recognition
- probabilistic neural network
- hidden markov models
- spectral analysis
- non stationary
- computer vision
- feature extraction
- feature vectors